RAC going steady down
Message boards :
Number crunching :
RAC going steady down
Message board moderation
| Author | Message |
|---|---|
|
Send message Joined: 6 Feb 10 Posts: 38 Credit: 274,204,838 RAC: 0 Level  Scientific publications   
|
Since the 24th of this month my RAC has gone down. I noticed that I am receiving only long runs with CUDA 42 when I was receiving before CUDA 55. I still think that the driver conflict between the Tesla K40 and the Titans is the cause because BOINC thinks I have four Titans when I actually have two Tesla and two Titans installed. The Tesla driver overrides the Titan driver. I will separate the Tesla from the Titans and see what happens. |
 Retvari Zoltan Send message Joined: 20 Jan 09 Posts: 2380 Credit: 16,897,957,044 RAC: 0 Level  Scientific publications  
|
You should upgrade your drivers to at least to 334.21. |
 Send message Joined: 17 Aug 08 Posts: 2705 Credit: 1,311,122,549 RAC: 0 Level  Scientific publications |
... is the cause because BOINC thinks I have four Titans when I actually have two Tesla and two Titans installed. This is actually just a reporting issue on the homepage, i.e. the web server code. The local BOINC should identify and handle different GPUs alright. You can see it in the message log right after startup. MrS Scanning for our furry friends since Jan 2002 |
|
Send message Joined: 6 Feb 10 Posts: 38 Credit: 274,204,838 RAC: 0 Level Scientific publications
|
The status event log states that all cards are using the same driver, the Tesla driver. It does show two Tesla and two Titans but all using the same driver 332.50 |
skgiven Send message Joined: 23 Apr 09 Posts: 3968 Credit: 1,995,359,260 RAC: 0 Level  Scientific publications  |
As already suggested, a driver update is in order. I suggest trying the latest Beta 337.12, and then moving to the next WHQL when it arrives... FAQ's HOW TO: - Opt out of Beta Tests - Ask for Help |
©2026 Universitat Pompeu Fabra